Economics of the bad cite

Citation

Pedersen, W.A. (2001), "Economics of the bad cite", The Bottom Line, Vol. 14 No. 4. https://doi.org/10.1108/bl.2001.17014daf.001

Economics of the bad cite

Economics of the bad cite

Keywords: Costs, Reference libraries, Bibliographic standards

Abstract

Incorrect, or "bad", references to the published literature have posed a long-standing problem to scholars, scientists and researchers, as well as librarians. The occurrence of bad cites in published bibliographies has been shown to range from 7 to 50 percent of listed citations. Little has been mentioned, however, about the adverse effects of these references, especially on interlibrary loan transactions. It is generally assumed that they are a nuisance, taking additional time and resources to correct and locate the desired document. Attempts to estimate the cost of bad cites to libraries using a combination of published reports and unpublished research findings. A rough price of $72.90 was determined. Suggestions for eliminating and reducing the problems associated with bad cites are presented along with suggestions for future research into this important topic.

Common to much of the world's scientific and scholarly literature is a particularly bothersome bibliographic entity that vexes researchers across all fields of study, including Library Science and Medicine (Benning and Speer, 1993). This problem is so commonplace – ranging from 7 percent to as high as 50 percent of the citations studied – that a review of the topic has been published (Sweetland, 1989). There are many names for this phenomenon: incorrect reference, bad cite, incorrect citation, incomplete reference, bibliographic error, incomplete citation, etc. For some people, these bad cites are merely irksome inconveniences. For others they are time-consuming and infuriating detours to research and scholarship. Yet incorrect citations are more than just personal challenges. They also represent an economic problem with consequences that can be measured in dollars and cents.

In any attempt to verify and correct a bad citation, there are certainly costs to be borne by the individual researcher in terms of their time. The more time an individual devotes to correcting one of these mistakes, the more expensive the process. Theoretically, the higher the salary of the person involved, the higher the cost to reconcile the problem cite. Many of these incorrect references end up at a library reference desk, where additional costs borne by the library are added to the inflationary cycle. In 1980 dollars, the cost of verifying a citation by reference staff at the National Library of Medicine was $3.37. Adjusted to 1996 dollars, the process of citation verification was calculated to cost $9.94 simply in terms of reference support (Abels, 1997).

Another place in the library where a bad cite often manifests itself is interlibrary loan (ILL). The vast majority of interlibrary loan requests come into the office and move through the workflow with few problems. Others, such as requests with incorrect citations, typically move from lower paid staff to higher paid staff, until the problem cite is corrected and the material can be ordered from another library (many times it ends up being available right there in the local library). One can speculate that the cost per transaction for this small group of difficult requests is higher than for routine requests.

The costs of interlibrary loan are well established in the library literature. In fact, ILL cost studies go all the way back to the depression (Hand, 1930). The average cost of an interlibrary loan-borrowing request (where a library obtains a book or document not owned locally from another library, document delivery company or other supplier) in 1996 dollars was reported by the Association of Research Libraries (ARL) to be $18.35 (Jackson, 1998). The same study reported the average cost of an interlibrary loan-lending request (where a library sends out a photocopy or returnable, such as a book, to another library) to be $9.48. Both of these figures are only averages and include all the routine requests as well as the less frequently encountered difficult requests, such as bad citations.

Cost data such as these are helpful, but research in the area of ILL costs needs to move beyond the concept of average costs to focus on specific categories of requests, such as bad cites. Zhou (1999) recognized the limitations of only calculating the overall mean when identifying interlibrary loan costs. New research needs to focus more on a detailed breakdown of costs for specific types of ILL requests, including such sub-categories as:

• •

  OCLC requests;

• •

  "rush" requests;

• •

  Docline requests;

• •

  requests that require the payment of document delivery or book loan fees to suppliers; and

• •

  those with difficult, or incorrect, citations.

The averaging of networking fees and document supply fees into ILL costs is a case in point. Like most interlibrary loan cost studies, the ARL study averaged networking fees, such as OCLC charges, across all requests, even those where the request was faxed or sent via another network service. Similarly, the cost of procuring documents and loans from other libraries and organizations was averaged across all requests, even though many of these requests were filled at no charge by the lender.

In 1990, while at the University of Texas Health Sciences Center at San Antonio (UTHSCSA), this author conducted an internal, unpublished cost study that allowed a more detailed breakdown of ILL costs and documented the higher costs associated with incorrect references. The methodology for this study was based upon an article written by Vinson (1988). All seven categories of costs examined in the 1996 ARL study were also reported in the 1990 UTHSCSA study: labor and benefits, network and communications costs, delivery costs, photocopy costs, supplies, borrower fees, equipment/software maintenance. Similar to ARL, the costs of a copy (non-returnable) and loan (returnable) were calculated separately.

In the UTHSCSA study, the average cost of obtaining a photocopied document was determined to be $4.79 and obtaining a returnable, such as a book, was $4.85. There are clearly a number of differences between the UTHSCSA and ARL studies. These include:

• •

  large cost differential;

• •

  size of the sample;

• •

  date of the data;

• •

  multi-institutional study versus single institution; and

• •

  research libraries versus academic health science center library.

Another difference between the two studies is that specific categories of requests were tracked and reported separately in the UTHSCSA study. For example, "rush" requests were found to actually cost less than regular requests: $3.87; networking fees, specifically OCLC costs, were applied only to those requests that utilized OCLC; and "difficult" requests, such as bad cites, were tracked separately. Specific categories of requests such as these were not detailed in the ARL study.

Procuring a document with an incorrect reference in the UTHSCSA study was determined to be $15.23; over three times (318 percent) the $4.79 cost of a routine document order. The additional cost was primarily attributable to the additional labor costs of more highly paid staff and computer searching fees. If one were to multiply the $18.35 average cost of a borrowing request reported by ARL in 1996 (this should be representative of the cost of a routine request with a good cite, because they greatly outnumber difficult ILL requests) by the multiple of three noted in the 1990 UTHSCSA study, the cost of a bad cite, simply in terms of interlibrary loan service, would be $55.05. This cost does not include the services of a reference librarian (often a library user will stop at the reference desk before going to interlibrary loan), nor does it include the cost of personnel at any other library service desk.

Using the figures reported above, the potential library cost for a single bad cite, if presented at both of these service points, would look something like Table I in 1996 dollars. Translated into year 2001 dollars, the price inflates to $72.90 (American Institute of Economic Research, 2001). This is by no means an exact figure; rather, it is a rough estimate based upon available data. If the true cost of a bad cite is anywhere near this $73 figure, the price of library support is quite high. And this does not include the costs borne by the individual researcher, academic department, and profit or non-profit organization. Considering that a single, published bad cite may be referenced and searched by hundreds, or even thousands, of people around the world for years to come, the scope of the economic problem goes far beyond $72.90.

Table I Potential library cost for a single bad cite

What can or should be done about this situation? A multi-faceted approach to decreasing the occurrence of bad cites is warranted:

• •

  Upgrade the bibliographic standards required by scholarly publications. Perhaps it is time for a universal adoption of the American National Standards Institute (1977) and a standardization of bibliographic citation formats across disciplines (Geer, 1995).

• •

  Publishers should adopt stringent guidelines for verifying bibliographic citations and move away from the use of abbreviations in bibliographic references. Often a bad cite is the result of a reader incorrectly guessing the full title of an abbreviation.

• •

  Indexing and abstracting services should also move away from the use of abbreviations in their products. This has been a particular problem in the engineering literature with its heavy use of conference proceedings, and with citations gleaned from Medline, the major index in the field of medicine. The publication of a National Standards Information Organization (NISO) standard (Z39.82-2001) on Title Pages for Conference Publications (NISO, 2001) should go a long way towards ameliorating difficulties with conference proceedings. Another positive development is that at least one version of Medline, specifically the OCLC Firstsearch version, now lists the full title of the serial along with the abbreviation.

• •

  Improve the training of researchers from all disciplines in the accurate reporting and interpretation of bibliographic citations. This type of training should be incorporated into the curriculum of all fields of study and also be offered in libraries through bibliographic instruction classes.

• •

  New interlibrary loan cost studies need to move beyond the concept of average cost and explore the costs of specific subcategories of requests, especially bad cites. These requests take an inordinate percentage of the total resources dedicated to ILL service and therefore deserve much more attention. Particular attention should be paid to testing the premise presented here that bad cites cost three times that of good references in terms of library support.

• •

  Current research dedicated to bad cites has focused heavily upon the rate of occurrence. Researchers should refocus their attention towards the deleterious effects of this problem for both library staff and individual researchers; particularly the economic costs.

In 1998, Rebaza spoke for librarians around the world when she wrote: "Although as librarians we have no control over what individual faculty choose to do, or what professional associations want to promote, it does seem to me that we should stand up for our patrons and lobby for an end to this sort of headache" (Rebaza, 1998). This is a problem that is going to take more than librarians to resolve, however. It will require the concerted efforts of everyone in the scholarly communication network, including authors, publishers, indexing and abstracting companies and professional organizations.

Everyone involved in the scholarly communication system is susceptible to the economic consequences of the bad cite. The very same people have a great deal to gain by assuring the accuracy and completeness of bibliographic citations before they go to print. Ascribing a cost of $72.90 to the average bad cite hopefully has an effect similar to the publication of average interlibrary loan cost data; namely, an increased awareness of the problem and widespread action to improve the situation. Until something substantive is done, there is little doubt that the cost of these bibliographic errors will continue to reverberate through time and space, costing ever more for future generations of researchers and librarians.

Wayne A. PedersenAssociate Professor, Head, Interlibrary Loan and Document Delivery, Iowa State University Library, Ames, Iowa, USA

References

Abels, E.G. (1997), "Improving reference service cost studies", Library and Information Science Research, Vol. 19 No. 2, pp. 135-52.

American Institute of Economic Research (2001), AIER Cost of Living Calculator, (Web document), The Institute, Chicago, IL. Available at http://www.aier.org/colcalc.html

American National Standards Institute (1977), American National Standard for Bibliographic References, American National Standards Institute, New York, NY.

Benning, S.P. and Speer, S.C. (1993), "Incorrect citations: a comparison of library literature with medical literature", Bulletin of the Medical Library Association, Vol. 81 No. 1, pp. 56-8.

Geer, B. (1995), "Unusual citings: journal citation integrity and the public services librarian", RQ, Vol. 35 No. 1, pp. 67-73.

Hand, E. (1930), "A cost survey in a university library", Library Journal, Vol. 55 No. 17, pp. 763-6.

Jackson, M. (1998), Measuring the Performance of Interlibrary Loan Operations in North American Research and College Libraries, Association of Research Libraries, Washington, DC.

National Information Standards Organization (NISO) (2001), Title Pages for Conference Publications, NISO Press, Bethesda, MD.

Rebaza, C. (1998), "When is a citation just a frustration?", College and Research Libraries News, Vol. 59 No. 7, pp. 512-13.

Sweetland, J.H. (1989), "Errors in bibliographic citations: a continuing problem", Library Quarterly, Vol. 59 No. 4, pp. 291-304.

Vinson, M. (1988), "Cost finding: a step-by-step guide", The Bottom Line: Managing Library Finances, Vol. 2 No. 3, pp. 15-19.

Zhou, J.Z. (1999), "Interlibrary loan cost studies and copyright fees", Journal of Interlibrary Loan, Document Delivery and Information Supply, Vol. 9 No. 4, pp. 29-38.